Making endo-cyclizations favorable again: A conceptually new synthetic approach to benzotriazoles via azide group directed lithiation/cyclization of 2-azidoaryl bromides

Article abstract of DOI:10.1039/c9ob00615j

Although benzotriazoles are important and ubiquitous, currently there is only one conceptual approach to their synthesis: bridging the two ortho-amino groups with an electrophilic nitrogen atom. Herein, we disclose a new practical alternative-the endo-cyclization of 2-azidoaryl lithiums obtained in situ from 2-azido-aryl bromides. The scope of the reaction is illustrated using twenty-four examples with a variety of alkyl, alkoxy, perfluoroalkyl, and halogen substituents. We found that the directing effect of the azide group allows selective metal-halogen exchange in aryl azides containing several bromine atoms. Furthermore, (2-bromophenyl)diazomethane undergoes similar cyclization to give an indazole. Thus, cyclizations of aryl lithiums containing an ortho-X = Y = Z group emerge as a new general approach for the synthesis of aromatic heterocycles. DFT computations suggested that the observed endo-selectivity applies to the anionic cyclizations of other functionalities that undergo "1,1-additions" (i.e., azides, diazo compounds, and isonitriles). In contrast, cyclizations with the heteroatomic functionalities that follow the "1,2-addition" pattern (cyanates, thiocyanates, isocyanates, isothiocyanates, and nitriles) prefer the exo-cyclization path. Hence, such reactions expand the current understanding of stereoelectronic factors in anionic cyclizations.

Full text of DOI:10.1039/c9ob00615j

View Article Online
View Journal
Organic &
Biomolecular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: A. Ageshina, G.
Chesnokov, M. Topchiy, I. Alabugin, M. Nechaev and A. Asachenko, Org. Biomol. Chem., 2019, DOI:
10.1039/C9OB00615J.
This is an Accepted Manuscript, which has been through the
Volume 14 Number
1 7 January 2016 Pages 1–372
Royal Society of Chemistry peer review process and has been
accepted for publication.
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1477-0520
COMMUNICATION
Takeharu Haino et al.
Solvent-induced emission of organogels based on tris(phenylisoxazolyl)
benzene
rsc.li/obc

Page 1 of 21
Organic & Biomolecular Chemistry
Making Endo-Cyclizations Favorable Again: Conceptually New
View Article Online
DOI: 10.1039/C9OB00615J
Synthetic Approach to Benzotriazoles via Azide Group Directed
Lithiation/Cyclization of 2-Azidoaryl Bromides
Alexandra A. Ageshina,[a] Gleb A. Chesnokov,[a,b] Maxim A. Topchiy,[a,b] Igor V.
Alabugin,[c] Mikhail S. Nechaev,[a,b] Andrey F. Asachenko[a,b]
[a] A.V. Topchiev Institute of Petrochemical Synthesis, Russian Academy of Sciences, 119991
Moscow, Leninskiy prospect 29, Russian Federation
[b] M.V. Lomonosov Moscow State University, 119991 Moscow, Leninskie Gory 1 (3), Russian
Federation
[c] Department of Chemistry & Biochemistry, Florida State University, 95 Chieftan Way,
Tallahassee, FL, 32306, USA
Abstract
Although benzotriazoles are important and ubiquitous, currently there is only one conceptual
approach to their synthesis: bridging the two ortho-amino groups with an electrophilic nitrogen.
Herein, we disclose a new practical alternative – the endo-cyclization of 2-azidoaryl lithiums
obtained in situ from 2-azido-aryl bromides. The scope of the reaction is illustrated by twenty-four
examples with a variety of alkyl, alkoxy, perfluoroalkyl, and halogen substituents. We found that
the directing effect of the azide group allows selective metal–halogen exchange in aryl azides
containing several bromine atoms. Furthermore, (2-bromophenyl)diazomethane undergoes similar
cyclization to give indazole. Thus, cyclizations of aryl lithiums containing an ortho–X=Y=Z group
emerge as a new general approach for synthesis of aromatic heterocycles.
DFT computations suggested that the observed endo-selectivity applies to the anionic
cyclizations of other functionalities that undergo the “1,1-additions” (i.e., azides, diazo compounds,
and isonitriles). In contrast, cyclizations with the heteroatomic functionalities that follow the “1,2-
addition” pattern (cyanates, thiocyanates isocyanates, isothiocyanates, nitriles) prefer the exo-
cyclization path. Hence, such reaction expands the current understanding of stereoelectronic factors
in anionic cyclizations.

Organic & Biomolecular Chemistry
Page 2 of 21
Introduction
View Article Online
DOI: 10.1039/C9OB00615J
The importance of cyclization reactions is underscored by the fact that the majority of
molecules in nature are cyclic.1 The success of a multi-step synthesis often depends on making the
key cyclic structure precisely and efficiently. Conceptually, classic nucleophilic cyclizations can
follow either an exo- or an endo-pathway.2 The endo-path, where the breaking bond is inside of the
forming cycle, is intrinsically unfavorable due to stereoelectronic penalty associated with the
unfavorable orbital overlap of the nucleophile with either * or * of the breaking bond.3 Hence,
the design of selective nucleophilic endo-cyclizations remains an important conceptual challenge.
The present manuscript shows that such challenge can be addressed by using the family of
functional groups that undergo “1,1-addition” reaction (Scheme 1). High endo-selectivity in these
reactions contrasts the usual preference for exo-cyclizations of the more common functionalities,
e.g., alkenes and alkynes, that undergo “1,2-additions”. In this work, we will explore the utility of
selective endo-cyclizations in synthesis of benzotriazoles.
Scheme 1. Comparison of “1,1-additions” and “1,2-additions“ in cyclization reactions. See
Scheme 11 for additional details
Benzotriazoles (Bts) find a wide range of applications in medicinal chemistry as illustrated
by such drugs as Vorozole and Alizapride.4 The search for the new Bt-based drugs continues - a
series of Bt derivatives have been identified as inhibitors of various kinases,5 inactivators of severe
acute respiratory syndrome 3CL protease,6 photoactivated DNA cleaving agents7 and agonists of
human orphan G-protein-coupled receptor GPR109b.8

Page 3 of 21
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00615J
Besides, many Bt derivatives are important in pharmacology since they exhibit
13
antidepressant,9 antimicrobial,10 antiviral,11 antimalarial,12 anti-inflammatory,4,
and
antimycobacterial activity.14 Furthermore, benzotriazoles are versatile synthetic auxiliaries in
diverse transformations15 and ligands in cross-coupling reactions.16
Today, there are only a few general methods to synthesize N-unsubstituted benzotriazoles.
The most common one is diazotization of ortho-phenylenediamines.17 Despite its frequent use, the
method has a number of drawbacks caused by limited availability and stability of substituted ortho-
phenylenediamines.
In all of the methods mentioned above, the 1,2,3-triazole ring of a benzotriazole is
assembled from an ortho-disubstituted benzene containing amino groups or its synthetic precursor
(Scheme 2).
Scheme 2. Comparison of the traditional and the new approaches to N-unsubstituted
benzotriazoles
Here, we present a new general synthetic approach to N-unsubstituted Bts from ortho-
metalated aryl azides. A unique feature of our approach is the unprecedented in situ transformation
of 2-azidophenyllithium into an N-lithium derivative of an unsubstituted benzotriazole. To the best
of our knowledge, no cyclization of the kind has been reported in literature. Proposed reaction
conditions are tolerant to a number of functional groups. Furthermore, synthetic approaches to the
requisite 2-azidoaryl bromides (iodides) allow access to substitution patterns that are different and
complementary to those available for ortho-phenylenediamines and their synthetic analogues (i.e.,
ortho-halogenated nitrobenzenes, ortho-nitroanilines, etc.). As the result, this method makes it
possible to synthesize benzotriazoles that have been unavailable through the previously reported
synthetic routes (vide infra).

Organic & Biomolecular Chemistry
Page 4 of 21
Results and Discussion
View Article Online
DOI: 10.1039/C9OB00615J
Azide cyclization
During the search for cyclization conditions, we tested different experimental approaches
using 2-azidobromobenzene as a model compound (Scheme 3). The attempt to obtain 2-
azidophenylmagnesium bromide by an excess of activated magnesium in refluxing THF failed; the
initial azide was regenerated in 92% yield. On the other hand, halogen-lithium exchange with n-
BuLi or t-BuLi in THF at –80 °С for 2h turned out to be more successful. Although application of t-
BuLi was accompanied by formation of intensely colored side products, treatment of the reaction
mixture with dilute hydrochloric acid resulted in the 62% yield of the benzotriazole. n-BuLi turned
out to be the most appropriate reagent providing 66% yield of the target benzotriazole.
Scheme 3. The one-pot formation of benzotriazole via lithiation of 2-azidobromobenzene
Inspired by these results, we decided to test the scope and limitations of the lithiation-
mediated cyclization with a wide range of readily available 2-azidobromobenzenes with various
substituents compatible with Li-organic reagents.
Substituted 2-azidobromoarenes were synthesized following a two-step approach (Scheme
4): bromination of substituted arylamines with N-bromosuccinimide in dichloromethane allowed 2-
bromoanilines 1 with yields 55–99%. Diazotization of 2-bromoanilines was performed according to
one of the two ways. The easily soluble anilines were diazotized under standard conditions in
aqueous solution of sulfuric acid,18 whereas polyhalogenated anilines and fluorine-containing
anilines were diazotized under water-free conditions, in trifluoroacetic acid.19
Scheme 4. General synthetic strategy for the preparation of substituted 2-azidobromoarenes
We found that 2-azidobromobenzenes containing alkyl substituents readily cyclized to
provide 1H-benzotriazoles 3b and 3c in excellent isolated yields (Scheme 5). Cyclization of 1-
azido-2-iodo-4-methylbenzene led to benzotriazole 3b in a lower yield than in case of the bromo
analogue. In the case of the iodo aryl precursor, isolation of the product was hampered by formation

Page 5 of 21
Organic & Biomolecular Chemistry
of the N-butylated benzotriazole byproduct. In the course of metal–halogen exchange, an
View Article Online
DOI: 10.1039/C9OB00615J
aryllithium derivative and corresponding n-butyl halide were formed. Under the reaction conditions,
1-bromobutane, formed from 2-azidobromoaryls, does not alkylate the lithium benzotriazole salt
formed in the course of the reaction. In contrast, the more reactive 1-iodobutane produces small
amounts of benzotriazole N-alkylation products (at N atoms 1 and 2).
Proposed method allows to synthesize polycyclic triazoles as demonstrated by the facile
formation of naphthotriazole 3d after Br-Li exchange of 2-bromo-1-azidonaphthalene 2d. The
tricyclic 3H-naphtho[1,2-d][1,2,3]triazole product was isolated in 84% yield.
1H-Benzotriazoles 3j–o from fluorine-containing aryl azides were obtained in good yields.
Scheme 5. Synthesis of benzotriazoles from 2-azidobromoaryls.[a]
[a] Reaction conditions: 2-azidoaryl bromide (5 mmol), 2.5M solution of n-BuLi (2.0 ml, 1.0 eq.) , THF (20 ml), –80
°C, 2h. Yields of isolated pure products are reported. [b] Cyclization 1-azido-2-iodo-4-methylbenzene. [c] For the
reaction, 40 ml THF was used.
The presence of two bromine atoms at the ortho-position with respect to the azido group
provided no complications - the bromo-substituted cyclic products 3g–i, 3n and 3o were isolated in
good yields (Scheme 5). Therefore, we decided to study the possibility of selective synthesis of

Organic & Biomolecular Chemistry
Page 6 of 21
benzotriazoles from polybrominated azides. A series of nine bromo-substituted benzotriazoles was
View Article Online
DOI: 10.1039/C9OB00615J
synthesized (Scheme 6).
[a] Reaction conditions: 2-azidoaryl bromide(5 mmol), 2.5M solution of n-BuLi (2.0 ml, 1.0 eq.) , THF (20 ml), –80 °C,
2h.
Scheme 6. Haloselective metal–halogen exchange in brominated aryl azides.[a]
Despite low solubility of alkyl-substituted azides 2u–w under the reaction conditions at –
80 °C, relevant benzotriazoles 3u–w were isolated in high yields. We found that the lithium–
halogen exchange is halogen- and regioselective in the presence of the ortho-azido substituent. Such
selectivity allows to synthesize a number of previously unknown substituted bromobenzotriazoles
in excellent yields. In cases when two halogens, bromine and chlorine, are located in ortho-
positions, metal–halogen exchange occurs only at bromine. For the di- and tribromosubstituted
substrates, only the bromine atom in ortho-position with respect to azide group was substituted with
lithium (Scheme 6).
It should be noted that because of the prototropic tautomerism of H-benzotriazoles,5b, 20
measurement of their 13C NMR spectra was complicated by considerable signal broadening. The
problem was solved by registering the NMR spectra in a 5:1 (by volume) mixture of DMSO-d6 and
D2O, with the addition of excess of NaOH (15 equivalents with respect to a benzotriazole).
Therefore, we showed that the azide moiety can serve as a directing group in lithiation
reactions. To the best of our knowledge, this directing effect has not been demonstrated previously
in the literature. It stems from the presence of unpaired electrons at the nitrogen atoms of the azide
group, which can coordinate the Lewis-acidic Li cation.

Page 7 of 21
Organic & Biomolecular Chemistry
Thus, the proposed method opens broad prospects for synthesis of new biologically active
View Article Online
DOI: 10.1039/C9OB00615J
substances. Many halogen-containing benzotriazoles are known to exhibit anticancer properties. For
example, 4,5,6,7-tetrabromo-1H-benzotriazole (TBBT) is a reference compound among CK2
inhibitors.21 Furthermore, the new method enables synthesis of benzotriazoles that could not be
accessed via conventional methods. For example, SciFinder® found no literature reports of ortho-
phenylenediamines needed for the synthesis of benzotriazoles 3v and 3w via diazotization.
The potential of the new bromobenzotriazoles in further functionalization via cross-coupling
reactions or via lithium-bromine exchange followed by treatment with electrophiles (DMF, CO2,
B(Oi-Pr)3, etc.), is obvious.
Indazole formation
Guided by the analogy between azides and other heteroatomic groups of –X=Y=Z type, we
have tested whether the ortho-lithiated benzenes substituted with such groups can undergo similar
cyclizations.
Indeed, we found that indazole synthesis can be accomplished using this method staring
from a diazo compound 4 (Scheme 7). 3-Phenylindazole 5 was isolated in a 28% yield along with
the 19% of N-butyl-3-phenylindazole 5a as a side-product.
Scheme 7. Cyclization of diazoderivative 4
The side-product 5a results from secondary reaction, alkylation of lithium salt 5 with 1-
bromobutane formed in the course of metal–halogen exchange. To avoid formation of 5a, we
substituted n-BuLi with t-BuLi. Treatment of diazocompound 4 with 2 equivalents22 of t-BuLi
resulted in 3-phenylindazole 5 being isolated in 29% yield. Interestingly, tert-butylated side-product
(N-tert-butyl-3-phenylindazole 5b) was isolated from a complex mixture in 6% yield.

Organic & Biomolecular Chemistry
Page 8 of 21
Theoretical analysis
View Article Online
DOI: 10.1039/C9OB00615J
Theoretical studies of the сyclization mechanisms were performed using DFT calculations
with the PBE functional.23 Valence electrons were treated using a TZ2P basis set, innermost
electrons of Li, C, N, O, S atoms were emulated using effective core potentials ECP-SBKJC.24
Stationary points were characterized as minima by calculations of normal modes of vibrations.
Total energies including zero-point vibration energy corrections were calculated E0 = E + ZPVE.
All calculations were made using PRIRODA program.25
In order to obtain realistic energies of the reactant, transition state and the product (lithiated
benzotriazole) for the reaction, we have included three molecules of THF coordinating to the Li
atom in our computational model (Scheme 8). As a reference point, we have also evaluated the
reaction energy profiles using the pure carbanion without the Li counterion and the solvent
molecules.
Scheme 8. Mechanism of cyclization of ortho-azido phenyl lithium (E0 in kcal/mol, relative to
the starting material).
According to the computational data, the cyclization of ortho-azido phenyl lithium to the
five-membered product RN3 is highly exothermic (-62.3 kcal/mol). Furthermore, the barrier for this
process is very low (5.6 kcal/mol).
From the nature of Frontier Molecular Orbitals (FMOs) of the reagent, the cyclization can be
described as a nucleophilic attack of the carbanionic lone pair (the HOMO) at the LUMO,
delocalized between the three azide nitrogens (Figure 1). Hence, one can draw an analogy of this
reaction with the known reactions of intermolecular nucleophilic addition of carbanions to organic
azides.26

Page 9 of 21
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00615J
HOMO (-3.92)
LUMO (-1.29)
Figure 1. The shape and energies of the frontier molecular orbitals of ortho-azido phenyl
lithium (orbital energies in eV)
The low calculated reaction barrier (5.6 kcal/mol) agrees well with the fast reaction
observed at the low temperatures used for the experimental metalation step. Considering how low
the barrier is, it is possible that the observed reaction rate mostly reflects the rate of metalation. To
further evaluate this scenario, we performed a series of experiments on lithium-halogen exchange of
2-azidobromobenzene followed by low temperature (-80 °C) treatment with various H-acids
(CH3OH, CF3COOH). In every case, we did not observe even traces of dehalogenated azides. Either
the 2-bromoazide reactant or the benzotriazole cyclization product was isolated depending on
duration of the lithium-bromine exchange.
In order to understand the role of cyclic constraints in the 5-endo TS, we compared the
energetic parameters of intramolecular reaction (Scheme 8) with the analogous intermolecular
addition of PhLi to PhN3 (Scheme 9).
Scheme 9. Mechanism of the intermolecular addition of PhLi to PhN3 (E0 in kcal/mol).
Comparison of reaction kinetics and thermodynamics provides important insights into the
nature of cyclization barrier. In particular, intermolecular nucleophilic addition is characterized by
even lower calculated energy barrier than in case if intramolecular reaction, 4.1 kcal/mol for TSN3-
inter vs. 5.6 kcal/mol for TSN3.

Organic & Biomolecular Chemistry
Page 10 of 21
Interestingly, the two reaction energies trends show an opposite trend than the activation
View Article Online
DOI: 10.1039/C9OB00615J
barriers. Although the product of the intermolecular reaction, the delocalized diphenyltriazenyl
lithium PN3-inter, is stabilized by double coordination of the lithium cation with both terminal
nitrogen centers, the exergonicity of the intermolecular reaction is lower (57.9 kcal/mol) in
comparison to the exergonicity of the intramolecular process (62.3. kcal/mol for PN3). Such
difference can be attributed to the greater stabilization associated the formation of the aromatic
triazole -system in the cyclization product PN3.
The intermolecular reaction also defines the general stereoelectronic requirement for the
nucleophilic attack at the azide moiety. In particular, it provides the preferred angle of attack at the
azide -system. This geometric parameter has been used extensively in analysis of reactivity and
selectivity of nucleophilic additions. For the alkene targets, such angle is called the Burgi-Dunitz
angle27 and usually lies in the region of 105-109 degrees. The angle of attack on alkynes has been
controversial: it was suggested to be acute in the classic work on the rules for cyclizations by
Baldwin2, 28 but later shown to be obtuse.29 Remarkably, the calculated angle of attack by PhLi at
the terminal nitrogen atom of phenyl azide (C-N-N = 106.1) is the same as the Burgi-Dunitz angle,
confirming that both geometric features originate from the generally preferred directionality for the
interaction of a donor at a *-orbital. The obtuse trajectory avoids interaction with the node
between the two atoms in the p-target. Exo-cyclizations where the breaking bond is outside of the
forming cycle can follow the obtuse trajectory and, thus, are generally stereoelectronically
preferred.
In contrast, the C-N-N angle for the 5-endo TS is much smaller (C-N-N = 72.4), providing a
possible explanation to why the intramolecular reaction has a higher barrier despite being more
exothermic.
Notably, the only reaction route that we were able to localize on the potential energy surface
corresponds to attack of phenyl anion to terminal nitrogen atom in TSN3-inter. No transition states
were found for reactions corresponding to attack on nitrogen atoms in medium position, or the one
attached to phenyl group. This is an important observation because it shows that 4-exo-cyclization
is not a viable option.
Intrigued by these results, we explored whether the cyclization of aryl lithium derivatives
with heteroatomic functionalities is general and whether it can be extended beyond the formation of
benzotriazoles from aryl azides. For this purpose, we have analyzed cyclizations of a number of
substrates with the general formula [Ph-X…Y…Z]Li(THF)3 (Scheme 10). In each case, we
considered both the 1,5-attack (5-endo) at the terminal atom Z and the 1,4-attack (4-exo) at the
central atom Y.

Page 11 of 21
Organic & Biomolecular Chemistry
View Article Online
DOI: 10.1039/C9OB00615J
Scheme 10. 1,5- vs 1,4-cyclizations in [Ph-X…Y…Z]Li(THF)3.
Before discussing the specific findings, we have to mention the specific differences between
the two main patterns for the addition of reactive species at a -system: i.e., 1,2-addition vs. 1,1-
addition. Such differences were clearly analyzed recently for radical additions to alkynes and
isonitriles.30 Although the two functionalities possess an analogous set of -orbitals that combine in
similarly looking HOMO and LUMO, the outcomes of their addition reactions are drastically
different. We will discuss these differences below by using anionic addition, pertinent to the present
discussion. Whereas addition to alkynes proceeds in a “1,2-manner” where the new bond and the
anionic center are formed at the different alkyne carbons, isonitriles can react as “1,1-synthons” by
forming both the new bond and the anionic center at the same terminal carbon of the isonitrile
moiety (Scheme 11). Such outcome is consistent with the hidden carbene nature of the isonitrile
functionality30 and accounts for the key role of isonitriles in multicomponent transformations such
as the Passerini and the Ugi reactions31 as well as the many interesting radical cascades of
isonitriles, sometimes described as an “insertion of isonitrile”.
Scheme 11. Comparison of polar “1,1-“, “1,2-“ and “1,3-additions” to selected functional
groups.
Although azides are electronically complex and have the potential of making either 1,1- or
1,3-products, the possibility of 1,1-products formation in addition to azides is the key feature that is

Organic & Biomolecular Chemistry
Page 12 of 21
similar to the 1,1-addition to isonitriles. The 1,2-addition to azides would give a charge-separated
View Article Online
DOI: 10.1039/C9OB00615J
zwitter-ionic product and is generally unfavorable. Also note that the “1,3” addition mode that is
possible in the intermolecular additions to azides, is more difficult in the presence of intramolecular
constraints in the o-Li-azidobenzene cyclization.
1,1-additions
As one could expect from the experimental preferences (vide infra), the preferred
transformation of the azide reactant RN3 is the formation of benzotriazole P1-5-N3 (Scheme 12). Not
only is the 1-4 activation barrier (~15 kcal/mol) is much higher but the putative product of the 1-4
reaction P1-4-N3 is 13.3 kcal/mol higher in energy than the reactant. Even if the 4-exo product were
kinetically competitive, its formation would be reversible. The fact that the 4-exo activation barrier
is only 2 kcal/mol higher than energy of the product suggests that this process is not
stereoelectronically unfavorable.
Scheme 12. [Ph-X…Y…Z]Li(THF)3 systems with preferred 1,1-additions (relative energies of
activation barriers E0 , intermediates and products E0 in kcal/mol, numbers in parenthesis
a
corresponds to reactions of anions in absence of Li+(THF)3).
a
no stationary points corresponding to transition states or products were found on potential
energy surface.
Considering the 1,1-pattern of addition to azides, one can also suggest that the 1-5 (endo)
preference is enhanced by the fact that the Li “counterion” does not need to migrate to an adjacent
atom as in the usual additions to alkenes and alkyne (the “1,2-additions”). In order to address this

Page 13 of 21
Organic & Biomolecular Chemistry
issue, we have also calculated the two cyclization potential energy profiles for the purely anionic
View Article Online
DOI: 10.1039/C9OB00615J
version of this reaction. As expected, the “naked” anion is more reactive and the 1-5 cyclization
barrier decreases noticeably (from 5.6 to 2.0 kcal/mol). Interestingly, the 1,4-product does not
correspond to an energy minimum in the absence of Li cation.
According to the computations, the diazo derivative RCN2 can be transformed in the
respective indazole P1-5-CN2 via negligible barrier of 0.5 kcal/mol and with high exothermicity (-62.3
kcal/mol). This finding suggested that the 1-5 reaction may be experimentally feasible. In contrast,
we could not locate a minimum at the potential energy surface that corresponds to the 1-4 product
P1-4-CN2. These findings agree well with the experimental observations.
In a similar manner, the endo-cyclization to an isonitrile proceeds via a low (1.8 kcal/mol)
barrier and is highly exothermic. Subsequent proton migration leads to a highly stable isoindole,
rendering the overall process to be 64.9 kcal/mol exergonic. The exo-cyclization does not provide a
product that corresponds to an energy minimum.
1,2-additions
For comparison, we have also investigated the reactions of isocyanate and isothicyanate
derivatives, RNCO and RNCS (Scheme 13). In these “1,2-additions” cases, exo-cyclizations become
the favorable direction. Polarization of the -bonds strongly favors nucleophilic attack at the
carbon. As the result, we could not find an energetically feasible path to the 5-endo-product from
the isocyanate. In contrast, the alternative 4-exo reaction has a low 8.0 kcal/mol activation barrier
and is sufficiently exothermic (E0 = -9.0 kcal/mol) to suggests that such four-membered structures
may be formed at the low temperature along with the products of their further transformations.
On the other hand, we have found both the 4-exo and the 5-endo products as energy minima
for the isothiocyanate substrate. In both cases, the cyclization process is exothermic (-7.9 and -19.8
kcal/mol). However, the reactions display relatively high barriers (27.9 and 17.3 kcal/mol,
respectively for P1-5-NCS and P1-4-NCS) and the intermolecular processes are likely to become
competitive because isothiocyanate groups are known to be highly reactive towards aryl lithiums.
We have also explored the transformation of cyanate ROCN and thiocyanate derivatives
RSCN. In both cases, the 5-endo-cyclizations have relatively low barriers (14.1 и 9.5 kcal/mol,
respectively) and are quite exergonic (-32.0 и -32.7 kcal/mol, respectively) because they lead to the
formation of aromatic benzoxazole P1-5-OCN and benzothiazole P1-5-SCN. However, the 4-exo-dig
cyclization was calculated to have even lower barriers (E0 = 5.6 и 5.3 kcal/mol, respectively).
a
Hence, the “normal”3a, 32 exo-preference for the “1,2-systems” is observed. Interestingly, the
formation of the four-membered ring is transient, and the reacting systems evolves further with the
migration of the cyano-group from the chalcogen to the ortho-carbon. The overall process results in

Organic & Biomolecular Chemistry
Page 14 of 21
the formation of the stable o-cyano phenolate P1-4-OCN and ortho-cyano thiophenolate P1-4-SCN, the
View Article Online
DOI: 10.1039/C9OB00615J
likely final products at these conditions.
Scheme 13. [Ph-N=C=Z]Li(THF)3 and [Ph-X-CN]Li(THF)3 systems with preferred 1,2-
additions (relative energies of activation barriers E0 , intermediates and products E0 in
a
kcal/mol, numbers in parenthesis corresponds to reactions of anions in absence of Li+(THF)3).
a
no stationary points corresponding to transition states or products were found on potential
energy surface.
An analogous kinetic preference for the 4-exo cyclization followed by a rearrangement was
observed for the reaction of benzyl cyanide RCCN (Scheme 14). The barrier for the cyclization is
only 8.8 kcal/mol. In this case, the 4-exo product corresponds to an intermediate P1-4-CCN. The latter
undergoes the C-C bond scission with the formation of ortho-cyano benzyl lithium via a low energy
(4.5 kcal/mol) transition state. The process is quite favorable thermodynamically -17.6 kcal/mol)
due to the strain relief and benzylic stabilization of the negative charge. The 5-endo product

Page 15 of 21
Organic & Biomolecular Chemistry
formation is much more favorable thermodynamically (-47.6 kcal/mol) but disfavored kinetically
View Article Online
DOI: 10.1039/C9OB00615J
(E0 = 16.4 kcal/mol).
a
Scheme 14. Computational analysis of cyclizations of [Ph-CH2-CZ]Li(THF)3 systems
following the 1,2-additions patterns (activation barriers E0 , relative energies of intermediates
a
and products E0 in kcal/mol, numbers in parenthesis corresponds to reactions of anions in
absence of Li+(THF)3).
Finally, we have analyzed reaction of the ortho-lithium derivative containing a non-
activated acetylene moiety RCCC. In the absence of activating electronegative heteroatoms, the 4-
exo barrier is significantly increased (27.1 kcal/mol). Much of this increase is due to the large
decrease of thermodynamic contribution to the reaction barrier:29b, 33 unlike the previous examples,
this 4-exo- product is a C-centered anion and, hence, reaction energy is significantly less favorable
(-6.2 kcal/mol).
On the other hand, the 5-endo barrier remains sufficiently low (16.3 kcal/mol) for this path
to remain kinetically viable at ambient temperatures. The initially formed indene P1-5-CCC-sigma (E0
= -38.0 kcal/mol) where the lithium is positioned at an sp2-hybridized carbon undergoes (E0 = -
a
14.5 kcal/mol) prototropic isomerization to a fully-conjugated aromatic anion P1-5-CCC-pi (E0 = -
65.3 kcal/mol). Hence, such reaction can be potentially a source of indenyl lithium derivatives of
this type.

Organic & Biomolecular Chemistry
Page 16 of 21
Conclusions
View Article Online
DOI: 10.1039/C9OB00615J
A conceptually new method of benzotriazole synthesis by intramolecular cyclization of
ortho-lithium aryl azides was developed. This method is based on robust lithium-halogen exchange
which is followed by fast and selective cyclization with azide is fast and selective. Furthermore, the
scope of the new method is broad - it was used to prepare a variety of polysubstituted
benzotriazoles including those that were previously unavailable.
From the general perspective, the newly developed approach for the construction of the five-
membered aromatic cycles opens synthetic access to four classes of heterocycles (benzotriazoles,
triazoles, indazoles, pyrazoles. In particular, despite the fact that benzotriazoles are mentioned in ca.
30 thousand references,1 This method was based on the introduction of the middle nitrogen atom to
N,N’-orthodisubstituted benzenes (Scheme 15). The new method is based on the creation of a C-N
bond from precursor that has the N3-moiety already installed.
The new method also works well for the construction of indazoles, another popular
heterocyclic scaffold as evident from ca. 15 thousands publications. In the future, the new method
may open synthetic avenues for the preparation of triazoles and pyrazoles, all of each are very
popular heterocycles.
Scheme 15
For the first time, directing effect of azide group in reactions of Li-Br exchange in aromatic
ring was demonstrated. This effect allows selective activation of o-Br substituents in the presence of
remote halogen atoms.
Computational analysis supports the possibility of intramolecular rearrangement at low
temperatures and provides insights into the selectivity of these processes. Computed geometries for
intermolecular addition to PhN3 revealed that the preferred stereoelectronic trajectory for the
1
The search for each heterocycle type was done using the following procedure – search for structures with at
least one reference, search for structures with at any experimental property, then “find all publications”

Page 17 of 21
Organic & Biomolecular Chemistry
nucleophilic attack at the azide group is similar to the classic Burgi-Dunitz trajectory for
View Article Online
DOI: 10.1039/C9OB00615J
nucleophilic addition to alkenes and alkynes.
However, the intramolecular reactions found dramatic differences between the functional
groups that undergo the two distinct addition patterns: the 1,1-additions (azides, isonitriles, diazo
compounds) and 1,2-additions (isocyanates, isothiocyanates, alkynes). Whereas exo-cyclizations are
preferred for the 1,2-additions, endo-cyclizations become more favorable for the 1,1-addition,
especially when the kinetic barrier is significantly lowered by thermodynamic contribution (e.g.,
from aromatic stabilization) and when the 4-exo-cyclizations are especially disfavored by the high
strain.
The new reaction opens broad possibilities for development of synthetic approaches to
heterocycles isoelectronic to benzotriazole through intramolecular cyclizations of the lithium–
phenyl–XYZ fragment. In particular, the theoretical predictions paved the way to the experimental
discovery of a new reaction, i.e., the indazole synthesis via cyclization of aryl-diazomethane lithium
derivatives. More detailed studies of the new process will be reported in the due course.
Acknowledgements
M. A. Topchiy, A. A. Ageshina, and A. F. Asachenko are thankful to Russian Science
Foundation (RSF) for financial support (project number 17-73-20023). Authors are grateful to the
Moscow State University (Russia) for the opportunity to use the NMR facilities of the Center for
Magnetic Tomography and Spectroscopy. Part of this work was carried out by M. S. Nechaev and
G. A. Chesnokov within the State Program of TIPS RAS.

Organic & Biomolecular Chemistry
Page 18 of 21
References
View Article Online
DOI: 10.1039/C9OB00615J
1.
2.
3.
R. S. Bon and H. Waldmann, Acc. Chem. Res., 2010, 43, 1103-1114.
J. E. Baldwin, J. Chem. Soc., Chem. Commun., 1976, 734-736.
a) I. V. Alabugin and K. Gilmore, Chem. Commun., 2013, 49, 11246-11250; b) G. P. Gomes
and I. V. Alabugin, J. Am. Chem. Soc., 2017, 139, 3406-3416.
4.
5.
I. Briguglio, S. Piras, P. Corona, E. Gavini, M. Nieddu, G. Boatto and A. Carta, Eur. J. Med.
Chem., 2015, 97, 612-648.
a) S.-O. Ka, H. P. Hwang, J.-H. Jang, I. Hyuk Bang, U.-J. Bae, H. C. Yu, B. H. Cho and B.-
H. Park, Sci. Rep., 2015, 5, 14816; b) R. Wąsik, M. Łebska, K. Felczak, J. Poznański and D.
Shugar, J. Phys. Chem. B, 2010, 114, 10601-10611; c) R. Swider, M. Masłyk, J. M. Zapico,
C. Coderch, R. Panchuk, N. Skorokhyd, A. Schnitzler, K. Niefind, B. de Pascual-Teresa and
A. Ramos, RSC Adv., 2015, 5, 72482-72494.
6.
C.-Y. Wu, K.-Y. King, C.-J. Kuo, J.-M. Fang, Y.-T. Wu, M.-Y. Ho, C.-L. Liao, J.-J. Shie,
P.-H. Liang and C.-H. Wong, Chem. Biol., 2006, 13, 261-268.
7.
a) P. A. Wender, S. M. Touami, C. Alayrac and U. C. Philipp, J. Am. Chem. Soc., 1996, 118,
6522-6523; b) J. W. Yang and P. A. Wender, Arch. Pharmacal Res., 1997, 20, 197-199.
G. Semple, P. J. Skinner, M. C. Cherrier, P. J. Webb, C. R. Sage, S. Y. Tamura, R. Chen, J.
G. Richman and D. T. Connolly, J. Med. Chem., 2006, 49, 1227-1230.
8.
9.
G. Caliendo, R. Di Carlo, R. Meli, E. Perissutti, V. Santagada, C. Silipo and A. Vittoria,
Eur. J. Med. Chem., 1993, 28, 969-974.
10.
a) N. A. Al-Awadi, N. M. Shuaib, A. Abbas, A. A. El-Sherif, A. El-Dissouky and E. Al-
Saleh, Bioinorg. Chem. Appl., 2008, 2008, 10; b) Z. Rezaei, S. Khabnadideh, K. Pakshir, Z.
Hossaini, F. Amiri and E. Assadpour, Eur. J. Med. Chem., 2009, 44, 3064-3067.
a) A. Carta, M. Loriga, S. Piras, G. Paglietti, M. Ferrone, M. Fermeglia, S. Pricl, P. Colla,
G. Collu, T. Sanna and R. Loddo, Med. Chem., 2007, 3, 520-532; b) M. Tonelli, G. Paglietti,
V. Boido, F. Sparatore, F. Marongiu, E. Marongiu, P. La Colla and R. Loddo, Chem.
Biodiversity, 2008, 5, 2386-2401; c) H. Gopi, M. Umashankara, V. Pirrone, J. LaLonde, N.
Madani, F. Tuzer, S. Baxter, I. Zentner, S. Cocklin, N. Jawanda, S. R. Miller, A. Schön, J.
C. Klein, E. Freire, F. C. Krebs, A. B. Smith, J. Sodroski and I. Chaiken, J. Med. Chem.,
2008, 51, 2638-2647.
11.
12.
13.
N. Mishra, P. Arora, B. Kumar, L. C. Mishra, A. Bhattacharya, S. K. Awasthi and V. K.
Bhasin, Eur. J. Med. Chem., 2008, 43, 1530-1535.
a) S. Bovens, M. Kaptur, A. S. Elfringhoff and M. Lehr, Bioorg. Med. Chem. Lett., 2009,
19, 2107-2111; b) A. Rajasekaran and K. A. Rajagopal, Acta Pharm. (Warsaw, Pol.), 2009,
59, 355-364.
14.
15.
a) A. Dubey, S. K. Srivastava and S. D. Srivastava, Bioorg. Med. Chem. Lett., 2011, 21,
569-573; b) A. Carta, M. Palomba, I. Briguglio, P. Corona, S. Piras, D. Jabes, P.
Guglierame, P. Molicotti and S. Zanetti, Eur. J. Med. Chem., 2011, 46, 320-326.
a) B. Zorc, Z. Rajić Džolić and I. Butula, Croat. Chem. Acta, 2012, 85, 595-602; b) B.
Chattopadhyay and V. Gevorgyan, Org. Lett., 2011, 13, 3746-3749; c) T. Horneff, S.
Chuprakov, N. Chernyak, V. Gevorgyan and V. V. Fokin, J. Am. Chem. Soc., 2008, 130,
14972-14974; d) S. Chuprakov, S. W. Kwok, L. Zhang, L. Lercher and V. V. Fokin, J. Am.
Chem. Soc., 2009, 131, 18034-18035; e) N. Grimster, L. Zhang and V. V. Fokin, J. Am.
Chem. Soc., 2010, 132, 2510-2511; f) B. Chattopadhyay and V. Gevorgyan, Angew. Chem.,
Int. Ed., 2012, 51, 862-872; g) S. Chuprakov, F. W. Hwang and V. Gevorgyan, Angew.
Chem., Int. Ed., 2007, 46, 4757-4759; h) S. Chuprakov, J. A. Malik, M. Zibinsky and V. V.
Fokin, J. Am. Chem. Soc., 2011, 133, 10352-10355; i) T. Miura, T. Biyajima, T. Fujii and
M. Murakami, J. Am. Chem. Soc., 2012, 134, 194-196; j) N. Selander, B. T. Worrell, S.
Chuprakov, S. Velaparthi and V. V. Fokin, J. Am. Chem. Soc., 2012, 134, 14670-14673; k)
A. R. Katritzky and S. Rachwal, Chem. Rev., 2010, 110, 1564-1610; l) Y. Wang, Y. Li, Y.
Fan, Z. Wang and Y. Tang, Chem. Commun., 2017, 53, 11873-11876; m) I. Alimi, R. Remy

Page 19 of 21
Organic & Biomolecular Chemistry
and C. G. Bochet, Eur. J. Org. Chem., 2017, 2017, 3197-3210; n) A. R. Katritzky and B. V.
View Article Online
Rogovoy, Chem. - Eur. J., 2003, 9, 4586-4593; o) A. R. Katritzky and X. Lan, Chem. Soc.
DOI: 10.1039/C9OB00615J
Rev., 1994, 23, 363-373.
16.
a) R. R. Jha, J. Singh, R. K. Tiwari and A. K. Verma, Arkivoc, 2012, 2013, 228; b) A. K.
Verma, in Adv. Heterocycl. Chem., ed. A. R. Katritzky, Academic Press, 2012, vol. 107, pp.
101-132; c) M. Singh, A. S. Singh, N. Mishra, A. K. Agrahari and V. K. Tiwari, ACS
Omega, 2019, 4, 2418-2424.
17.
a) R. E. Damschroder and W. D. Peterson, Org. Synth., 1940, 20, 16; b) S. Azeez, P.
Chaudhary, P. Sureshbabu, S. Sabiah and J. Kandasamy, Org. Biomol. Chem., 2018, 16,
6902-6907; c) Y. Luo, Z.-J. Liu, G. Chen, J. Shi, J.-R. Li and H.-L. Zhu, MedChemComm,
2016, 7, 263-271.
18.
19.
20.
21.
A. Cwiklicki and K. Rehse, Arch. Pharm., 2004, 337, 156-163.
K. Kanakarajan, K. Haider and A. W. Czarnik, Synthesis, 1988, 1988, 566-568.
J. Poznański, A. Najda, M. Bretner and D. Shugar, J. Phys. Chem. A, 2007, 111, 6501-6509.
a) R. Wąsik, P. Wińska, J. Poznański and D. Shugar, PLOS ONE, 2012, 7, e48898; b) R.
Battistutta, E. De Moliner, S. Sarno, G. Zanotti and L. A. Pinna, Protein Sci., 2001, 10,
2200-2206; c) S. Sarno, H. Reddy, F. Meggio, M. Ruzzene, S. P. Davies, A. Donella-Deana,
D. Shugar and L. A. Pinna, FEBS Letters, 2001, 496, 44-48; d) P. Zien, J. S. Duncan, J.
Skierski, M. Bretner, D. W. Litchfield and D. Shugar, Biochim. Biophys. Acta, Proteins
Proteomics, 2005, 1754, 271-280; e) M. Winiewska, K. Kucińska, M. Makowska, J.
Poznański and D. Shugar, Biochim. Biophys. Acta, Proteins Proteomics, 2015, 1854, 1708-
1717; f) Mario A. Pagano, J. Bain, Z. Kazimierczuk, S. Sarno, M. Ruzzene, G. Di Maira, M.
Elliott, A. Orzeszko, G. Cozza, F. Meggio and Lorenzo A. Pinna, Biochem. J., 2008, 415,
353-365.
22.
23.
a) E. J. Corey and D. J. Beames, J. Am. Chem. Soc., 1972, 94, 7210-7211; b) W. F. Bailey
and E. R. Punzalan, J. Org. Chem., 1990, 55, 5404-5406.
a) J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865-3868; b) M. G.
Medvedev, I. S. Bushmarinov, J. W. Sun, J. P. Perdew and K. A. Lyssenko, Science, 2017,
355, 49-52.
24.
a) W. J. Stevens, H. Basch and M. Krauss, J. Chem. Phys., 1984, 81, 6026-6033; b) W. J.
Stevens, M. Krauss, H. Basch and P. G. Jasien, Can. J. Chem.-Rev. Can. Chim., 1992, 70,
612-630; c) T. R. Cundari and W. J. Stevens, J. Chem. Phys., 1993, 98, 5555-5565.
a) D. N. Laikov, Chem. Phys. Lett., 1997, 281, 151-156; b) D. N. Laikov and Y. A.
Ustynyuk, Russ. Chem. Bull., 2005, 54, 820-826.
S. Brase, C. Gil, K. Knepper and V. Zimmermann, Angew. Chem.-Int. Edit., 2005, 44, 5188-
5240.
a) H. B. B:urgi, J. D. Dunitz, J. M. Lehn and G. Wipff, Tetrahedron, 1974, 30, 1563-1572;
b) N. K. Devaraj and C. L. Perrin, Chem. Sci., 2018, 9, 1789-1794.
J. E. Baldwin, R. C. Thomas, L. I. Kruse and L. Silberman, J. Org. Chem., 1977, 42, 3846-
3852.
a) K. N. Houk, N. G. Rondan, P. v. R. Schleyer, E. Kaufmann and T. Clark, J. Am. Chem.
Soc., 1985, 107, 2821-2823; b) I. V. Alabugin, K. Gilmore and M. Manoharan, J. Am.
Chem. Soc., 2011, 133, 12608-12623.
25.
26.
27.
28.
29.
30.
31.
G. P. Gomes, Y. Loginova, S. Z. Vatsadze and I. V. Alabugin, J. Am. Chem. Soc., 2018,
140, 14272-14288.
a) M. Passerini and L. Simone, Gazz. Chim. Ital., 1922, 52, 126; b) M. Passerini and L.
Simone, Gazz. Chim. Ital., 1922, 52, 181; c) I. Ugi and R. Meyr, Angew. Chem., 1958, 70,
702-703; d) I. Ugi, R. Meyr and C. Steinbrückner, Angew. Chem., 1959, 71, 386; e) F.
Bauer, H. Braunschweig and K. Schwab, Organometallics, 2010, 29, 934-938.
K. Gilmore, R. K. Mohamed and I. V. Alabugin, Wiley Interdiscip. Rev.: Comput. Mol. Sci.,
2016, 6, 487-514.
32.

Organic & Biomolecular Chemistry
Page 20 of 21
33.
For the more detailed discussion of connection between kinetics and thermodynamics in
View Article Online
cyclizations, see: a) K. Gilmore, M. Manoharan, J. I. C. Wu, P. v. R. Schleyer and I. V.
Alabugin, J. Am. Chem. Soc., 2012, 134, 10584-10594; b) S. F. VasileDvOsIk: 1y0,.10T39./Cy9O.B0F0.615J
Mikhailovskaya, V. I. Mamatyuk, G. E. Salnikov, G. A. Bogdanchikov, M. Manoharan and
I. V. Alabugin, J. Org. Chem., 2009, 74, 8106-8117; c) I. V. Alabugin and M. Manoharan,
J. Am. Chem. Soc., 2005, 127, 12583-12594; d) I. V. Alabugin and M. Manoharan, J. Am.
Chem. Soc., 2005, 127, 9534-9545.

Page 21 of 21
Organic & Biomolecular Chemistry
TOC entry
View Article Online
DOI: 10.1039/C9OB00615J